American Journal of Epidemiology
Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health 2011.
Vol. 174, No. 10
DOI: 10.1093/aje/kwr250
Advance Access publication:
October 7, 2011
Original Contribution
Prepregnancy Body Mass Index and Gestational Weight Gain in Relation to Child
Body Mass Index Among Siblings
Amy M. Branum*, Jennifer D. Parker, Sarah A. Keim, and Ashley H. Schempf
* Correspondence to Amy M. Branum, National Center for Health Statistics, Centers for Disease Control and Prevention,
3311 Toledo Road, Room 6113, Hyattsville, MD 20782 (e-mail: [email protected]).
Initially submitted February 15, 2011; accepted for publication June 28, 2011.
There is increasing evidence that in utero effects of excessive gestational weight gain may result in increased weight
in children; however, studies have not controlled for shared genetic or environmental factors between mothers and
children. Using 2,758 family groups from the Collaborative Perinatal Project, the authors examined the association of
maternal prepregnancy body mass index (BMI) and gestational weight gain on child BMI at age 4 years using both
conventional generalized estimating equations and fixed-effects models that account for shared familial factors. With
generalized estimating equations, prepregnancy BMI and gestational weight gain had similar associations with the
child BMI z score (b ¼ 0.09 units, 95% confidence interval (CI): 0.08, 0.11; and b ¼ 0.07 units, 95% CI: 0.04, 0.11,
respectively). However, fixed effects resulted in null associations for both prepregnancy BMI (b ¼ 0.03 units,
95% CI: 0.01, 0.07) and gestational weight gain (b ¼ 0.03 units, 95% CI: 0.02, 0.08) with child BMI z score at
age 4 years. The positive association between gestational weight gain and child BMI at age 4 years may be
explained by shared family characteristics (e.g., genetic, behavioral, and environmental factors) rather than in
utero programming. Future studies should continue to evaluate the relative roles of important familial and environmental factors that may influence BMI and obesity in children.
birth weight; body mass index; family; fixed effects model; pregnancy outcome; prenatal nutritional physiological
phenomena; weight gain
Abbreviations: BMI, body mass index; CI, confidence interval; CPP, Collaborative Perinatal Project; GEE, generalized estimating
equation.
In light of recent Institute of Medicine guideline revisions (1),
maternal prepregnancy weight and gestational weight gain are
receiving increasing attention as important risk factors for
adverse pregnancy and childhood outcomes. Specifically, overweight in children may be a potentially important consequence
of prepregnancy maternal obesity or excessive weight gain
during pregnancy through presumed in utero programming
of appetite, metabolism, or another biologic weight-related
mechanism (2). Although these potential relations have implications for intervention, there are many shared characteristics
between mother and child (e.g., genetic, environmental, behavioral) that may confound an observed intrauterine influence
of maternal body size or gestational weight gain on offspring
size. Nevertheless, the potential contribution of high pregnancy weight gain to childhood obesity remains a question of
interest since both measures have been increasing and they are
plausibly related through a fetal programming hypothesis.
An increasing number of studies have examined the possible
effects of maternal prepregnancy body mass index (BMI)
and/or gestational weight gain on offspring size in childhood,
with conflicting results, although most have found positive
associations (3–8). However, none of these studies was able
to control for confounding by genetic or environmental influences shared between mother and offspring. These shared
factors are important to consider as the general heritability
of BMI has been reported to range from 50% to 80% (9, 10),
and there is some evidence that shared and nonshared environmental factors are also important contributors to BMI during
childhood and later life (11–15). Knowledge of the source of
the association between maternal prepregnancy BMI and/or
1159
Am J Epidemiol. 2011;174(10):1159–1165
1160 Branum et al.
prenatal weight gain and child BMI carries important implications for informing interventions designed to reduce
child obesity. If the relation cannot be explained by shared
factors, then there may be a critical period of importance to
address the modifiable environmental/behavioral influences
of maternal weight; otherwise, there may be no deterministic impact of the prenatal period and, thus, the timing of
intervention.
Our primary objective was to examine the association of
maternal prepregnancy BMI and gestational weight gain on
child BMI at age 4 years by using a longitudinal sample of
pregnant women and their offspring to account for unobserved familial traits with a fixed-effects approach.
By contrasting only the perinatal exposures and outcomes
of siblings born to a given woman, the fixed-effects approach controls for shared family-level environmental, genetic, and behavioral characteristics that may be related to
both gestational weight gain and offspring size.
MATERIALS AND METHODS
Study population
We analyzed data from the Collaborative Perinatal Project
(CPP) that enrolled women at their first prenatal visit at 12 US
sites (1959–1965) and followed the offspring to the age of
7 years (16). The CPP is still frequently used to answer important questions of the perinatal and early childhood periods
as it remains the largest source of longitudinal data in the
United States on many pregnancy and childhood outcomes (17).
Women in the CPP who gave birth to term (37 weeks)
singleton infants with no preexisting or gestational diabetes
were eligible for our study. Mothers could enter the CPP at
any pregnancy, and study pregnancies were not necessarily
consecutive. This analysis was found to be exempt from institutional review board review by the National Institutes
of Health Office of Human Subjects Research and by the
National Center for Health Statistics.
Study sample and exclusions
There were 44,261 singleton term livebirths to nondiabetic
mothers in the original CPP data set eligible for our study.
About 65% of these children (n ¼ 28,746) had a follow-up
visit at age 4 years. Of these, 166 children had implausible
BMI values, according to the Centers for Disease Control and
Prevention algorithm (18), and another 219 children had information missing on either height or weight. Of the resulting
28,361 children, 2,074 were excluded because of gestational
weight gain measures that were either missing or could not be
reliably assigned to an obstetric visit as the result of errors
in dating the visits at the time the study was done. Finally,
a further 907 children were excluded because of missing data
on maternal height and another 354 because of missing
data on prepregnancy weight. This left a final sample size
of 19,109 ‘‘nonsiblings’’ and 5,917 individual children in
2,758 sibling groups for analysis.
Compared with the sample of eligible mothers prior to
exclusions for attrition, missing data, and nonsiblings, mothers
in the analytical sample were slightly less likely to be black.
Otherwise, the demographic characteristics were similar.
Information on 2 pregnancies was contributed by 87% of the
mothers, 12% contributed information on 3 pregnancies, and
1% of the mothers contributed information on 4 or more
pregnancies. The majority of the women (93%) contributed
information on at least 2 consecutive pregnancies. The average
age of mothers in the study sample was 24 years, and the
average parity was 2.
Exposure and outcome variables
Maternal prepregnancy BMI and gestational weight gain,
measured as continuous and categorical variables, were
the main exposures. Maternal prepregnancy BMI (weight
(kg)/height (m)2) was calculated on the basis of self-reported
prepregnancy weight recorded at enrollment and measured
height in inches without shoes. At each subsequent clinic
visit, women were weighed in street clothes without shoes
or a coat. Recognizing that there is no best measure of
weight gain during pregnancy (19), we calculated gestational
weight gain as measured weight at the last prenatal visit
within 3 weeks of delivery—prepregnancy weight. In addition
to continuous measures, categories of both weight gain and
prepregnancy BMI were also examined. For weight gain, the
current Institute of Medicine guidelines were followed, which
classify women as inadequate, adequate, or excess gainers
according to prepregnancy BMI (1). Prepregnancy BMI categories were based on current guidelines for weight status:
underweight ¼ BMI <18, normal weight ¼ BMI 18–<25,
overweight ¼ BMI 25–<30, and obese ¼ BMI 30.
The main outcome was child BMI-for-age z score, analyzed
as a continuous variable. Children were examined at 4 years
of age (mean age at visit, 48.1 months; standard deviation, 1.3).
Height was measured to the nearest 0.5 cm by using a standardized backboard, and weight was recorded in pounds to the
nearest 0.25 pound (113.39 g) or grams to the nearest 100 g
by use of scales calibrated semiannually. Gender-specific childhood BMI-for-age z scores were calculated on the basis of the
year 2000 Centers for Disease Control and Prevention growth
charts (20).
Covariates
Potential maternal demographic and pregnancy-specific
covariates identified a priori from previous studies included
maternal race, maternal socioeconomic status, maternal age,
parity, smoking during pregnancy, gestational age, birth
weight, and child’s sex. For each pregnancy, smoking, parity,
and maternal age information was collected at the first prenatal visit interview. Information on gestational age, calculated on the basis of the last menstrual period, birth weight,
recorded in grams of weight at delivery, and child’s sex was
collected from obstetric records. ‘‘Socioeconomic status’’
was defined as a continuous socioeconomic index developed
by the US Census Bureau that incorporated the educational
level and occupation of the head of household and family
income as reported at the first prenatal visit (21, 22).
Statistical analysis
We assessed linearity in the relation between gestational
weight gain and child BMI z score using spline regression
Am J Epidemiol. 2011;174(10):1159–1165
Gestational Weight Gain and Child BMI in Siblings
(23). After adjusting for prepregnancy BMI and graphically
examining this relation using locally weighted regression
to smooth scatterplots (lowess smoothing), we found an
apparently significant increase in the relation between gestational weight gain and child BMI z score for children
whose mothers gained more than 0 kg. However, a test of
differences in slopes at this cutpoint was not statistically
significant; therefore, the data were analyzed as a linear
relation.
We first performed conventional linear regression analysis
with generalized estimating equations (GEEs) to adjust standard errors for the clustering of siblings within families. The
results of these models are comparable to those of previous
studies that contrast different mother-child combinations without conditioning on the mother or shared family traits. Next,
we built family fixed-effects regression models to control for
all stable unmeasured variables that might be associated with
intrauterine factors and offspring size (e.g., maternal genetics,
environmental factors), by performing matched sibling contrasts within families (24–26). With fixed-effects models,
the inference reveals whether changes in maternal factors for
a particular mother, such as prepregnancy BMI or gestational
weight gain, are related to differences in offspring size at
4 years. By contrast, models without family fixed effects compare between or across mother-child combinations so associations do not ‘‘fix’’ or hold constant any familial factors that
are not observed and explicitly controlled.
Using the approach described above, we began by examining the association between prepregnancy BMI and gestational
weight gain on child BMI and considered the potentially
confounding or mediating role of infant birth weight on those
relations by presenting results with and without the inclusion
of birth weight.
For all models, adjustment was made for maternal race,
smoking during pregnancy (binary), maternal age (continuous),
socioeconomic index (categorical), parity (categorical), gestational age (continuous), and child’s sex (binary). Effect
estimates are presented per 5 kg for gestational weight gain,
per 2-unit change in BMI for maternal prepregnancy BMI,
and per 50 g of infant’s birth weight. A 2-unit change in BMI
for a woman corresponds approximately to a 5-kg change
in weight. We also assessed a potential interaction between
gestational weight gain and prepregnancy BMI on the outcome
of child BMI by introducing a cross-product term to the
model. All analyses were conducted in SAS, version 9.2,
software (SAS Institute, Inc., Cary, North Carolina) by using
PROC GENMOD with the ‘‘independent’’ correlation structure for GEE models and PROC GLM (with the ABSORB
statement) for fixed-effect models.
RESULTS
Socioeconomic status changed for at least 1 pregnancy
among 40% of the women; 49% of the mothers smoked during
1 or more pregnancies, and 7% changed smoking categories
across pregnancies (Table 1). Mothers averaged a prepregnancy
BMI of 22.9 (range, 13.7–52.2), and 24% changed prepregnancy BMI category between the first and second pregnancies in the sample (Table 2). Approximately 9 percent of
the women in this sample were classified as underweight,
Am J Epidemiol. 2011;174(10):1159–1165
1161
Table 1. Selected Demographic and Pregnancy Characteristics of
Study Mothers, National Collaborative Perinatal Project, 1959–1974
Mothers (n 5 2,758)
Characteristic
Mean (SD)
No. of pregnancies in study
Average paritya
Average maternal age, yearsa
No.
%
5,917
2.1 (2.0)
24.1 (5.3)
Race (n ¼ 2,758)
White
1,659
60.2
Black
1,028
37.3
Other
65
2.3
6
0.2
1,305
50.6
Missing
Smoking (n ¼ 2,578)
No pregnancies
Some but not all pregnancies
185
7.2
1,088
42.2
<33 (low)
634
23.0
33–46
620
22.5
47–62
626
22.7
63
827
29.9
All pregnancies
Socioeconomic index during
first pregnancy (n ¼ 2,758)
Missing
Socioeconomic status the same
category for all pregnancies
51
1.9
60.3
Abbreviation: SD, standard deviation.
These statistics are averaged for each mother and then averaged
among all mothers.
a
and 6 percent were classified as obese prior to their first
pregnancy in the sample by modern BMI standards. Mothers
averaged 9.9 kg of gestational weight gain, and 40% changed
gestational weight gain category between the first and second
pregnancies in the sample. Children in the sample averaged approximately 3,300 g at birth and had an average BMI
of 16 at age 4 years (Table 3). The majority of children (80%)
represented the first and second children included in the CPP
but represented the second and third or higher children in the
family.
According to GEE models, prepregnancy BMI and gestational weight gain had similar associations with child BMI
z score (b ¼ 0.09 units, 95% confidence interval (CI): 0.08,
0.11, and b ¼ 0.07 units, 95% CI: 0.04, 0.11, respectively)
(Table 4). However, neither prepregnancy BMI nor gestational weight gain was significantly predictive of child
BMI at age 4 years in fixed-effects models (b ¼ 0.02 units,
95% CI: 0.02, 0.06, and b ¼ 0.03 units, 95% CI: 0.03, 0.08,
respectively). Adjustment for birth weight resulted in attenuated GEEs and fixed-effects estimates, yet the estimate for
birth weight remained unchanged after adjustment for prepregnancy BMI and gestational weight gain. The interaction
term between prepregnancy BMI and gestational weight was
not significant for either type of model. Results by categorical
values of gestational weight gain adjusted for birth weight
1162 Branum et al.
Table 2. Descriptions of Gestational Weight Gain and Prepregnancy Body Mass Index Across the First 2 Pregnancies Represented in the Study
Sample, National Collaborative Perinatal Project, 1959–1974
Pregnancy 1
(n 5 2,758)
No.
Pregnancy 2
(n 5 2,758)
%
No.
Normal weight
262
9.5
212
7.7
1,944
70.5
1,848
67.0
Overweight
396
14.3
500
18.1
Obese
156
5.7
198
7.2
Pregnancy 1
(n ¼ 2,757)
Pregnancy 2
(n ¼ 2,758)
No.
%
No.
%
1,610
58.4
1,713
62.1
d
Gestational weight gain
Inadequate
Mean Prepregnancy
BMI (SD)b
23.9
22.9 (4.0)
% Changec
Mean Gestational
Weight Gain (SD), kg
Prepregnancy
BMI Range
%
Prepregnancy body mass index
Underweight
% Changea
40.4
Adequate
837
30.4
735
26.7
Excess gainers
310
11.2
310
11.2
13.7 to 52.2
Range of Gestational
Weight Gain, kg
9.9 (3.9)
12.7 to 34.9
Abbreviations: BMI, body mass index; SD, standard deviation.
Represents change in prepregnancy BMI category over 1 or more pregnancies represented in the study sample.
b
Means are averaged for each mother and then averaged over the entire sample.
c
Represents change in gestational weight gain category between the first and second pregnancies in the study sample.
d
The gestational weight gain category is also based on prepregnancy body mass index.
a
were similar to results using continuous measures (Table 5).
For categories of prepregnancy BMI, women who were overweight or obese prior to pregnancy had children with larger
BMI at age 4 years in GEE models but not fixed-effects models.
However, use of categorical variables demonstrated that
women gaining inadequate gestational weight had children with
greater BMI at age 4 years in fixed-effects models (b ¼ 0.08,
95% CI: 0.00, 0.16).
Table 3. Selected Characteristics of Study Siblings, National
Collaborative Perinatal Project, 1959–1974
Sibling Offspring (n 5 5,917)
Characteristic
Mean (SD)
Child BMI at age 4 years
Birth weight, g
No.
%
1,065
18.0
16.1 (1.4)
3,307 (472)
Birth order in family
First
Second
1,652
27.9
Third or more
3,121
52.8
79
1.3
Missing
Birth order in study
First
2,461
41.6
Second
2,579
43.6
Third
720
12.2
Fourth or more
157
2.6
2,947
49.8
Female child
Abbreviations: BMI, body mass index; SD, standard deviation.
DISCUSSION
Our objective was to examine the association between prepregnancy maternal size and gestational weight gain on subsequent child BMI within families in order to account for shared
factors, such as genetic propensity for size or family diet/
activity patterns, which may explain this relation. Unlike the
results using more standard regression techniques that do
not account for unmeasured family factors, the fixed-effects
analysis did not reveal a positive association between prepregnancy BMI and gestational weight gain and child BMI at
4 years of age. In fact, the within-family analysis suggested
that only inadequate weight gain rather than excess weight
gain may be associated with increased child BMI.
Our findings suggest that shared familial characteristics,
many of which previous studies had not measured, may explain the positive relation between gestational weight gain
and child BMI. For example, our GEE results from the continuous models were similar to at least one other study that
used more modern data but could not account for shared
family-level factors. Oken et al. (5) reported a 0.13 increase
in child BMI at age 3 years for every 5 kg of gestational
weight gain after adjusting for parental BMI. Our results
from the continuous model adjusted for birth weight were also
similar to those from 2 other studies that reported no association between gestational weight gain and increased fat mass
or overweight in children at the ages of 3 (27) and 4 (7) years,
after adjusting for birth weight. In addition, similar to our
findings from the GEE analysis, other studies have reported
a stronger effect of prepregnancy BMI on child size compared with gestational weight gain. Fisch et al. (3), using data
from only the Minnesota CPP site, did not find a significant
Am J Epidemiol. 2011;174(10):1159–1165
Gestational Weight Gain and Child BMI in Siblings
Table 4. Estimates of Change in Child BMI z Score at Age 4 Years
With Continuous Prepregnancy BMI and Gestational Weight Gain
Among Term Liveborn Siblings, National Collaborative Perinatal
Project, 1959–1974
Continuous
Measures
GEE
b
95% CI
Fixed Effects
b
95% CI
Model 1a
Prepregnancy
BMI
0.09b
0.08, 0.11
0.02b
0.02, 0.06
Gestational
weight gain
0.07c
0.04, 0.11
0.03c
0.03, 0.08
Model 2a
Prepregnancy
BMI
0.07b
0.06, 0.09
0.01b
0.05, 0.04
Gestational
weight gain
0.01c
0.02, 0.04
0.03c
0.08, 0.02
Birth weight
(per 50 g)
0.03d
0.02, 0.03
0.03d
0.02, 0.03
1163
Table 5. Estimates of Change in Child BMI z Score at Age 4 Years
With Categorical Prepregnancy BMI and Gestational Weight Gain
Among Term Liveborn Siblings, National Collaborative Perinatal
Project, 1959–1974
Categorical
Measures
GEE
b
95% CI
Fixed Effects
b
95% CI
Gestational
weight gaina,b
Inadequate
weight gain
0.04
Adequate
weight gain
1.00
Excess
weight gain
0.01
0.02, 0.10
Referent
0.08, 0.10
0.08
0.00, 0.16
1.00
Referent
0.01
0.13, 0.14
0.12
0.28, 0.08
Prepregnancy
BMIa
Underweight
0.27
0.38, 0.16
Normal weight
1.00
Referent
1.00
Overweight
0.16
0.09, 0.24
0.05
0.16, 0.08
Referent
Obese
0.23
0.12, 0.34
0.08
0.30, 0.14
Abbreviations: BMI, body mass index; CI, confidence interval; GEE,
generalized estimating equation.
a
Models were adjusted for maternal race, smoking, maternal age,
socioeconomic index, parity, gestational age, and child’s sex.
b
Estimate is change in age 4 BMI z score per 2-unit change in
prepregnancy BMI.
c
Estimate is change in age 4 BMI z score per 5-kg change in
gestational weight gain.
d
Estimate is change in age 4 BMI z score per 50-g change in birth
weight.
Abbreviations: BMI, body mass index; CI, confidence interval;
GEE, generalized estimating equation.
a
Models were further adjusted for maternal race, birth weight,
smoking, maternal age, socioeconomic index, parity, gestational
age, and child’s sex.
b
Gestational weight gain was based on Institute of Medicine categories of gestational weight gain and prepregnancy BMI.
association between gestational weight gain and child size at
age 4 or 7 years but did find a significant and positive association between the mother’s weight and height ratio index
(weight divided by height) at birth, age 4 years, and age
7 years. Others report only a significant association of gestational weight gain as modified by prepregnancy BMI (6, 8).
Our results from the fixed-effects analysis are consistent
with those from other research and observations regarding
genetic and ex utero environmental contributions to child
size. For example, there is emerging evidence that gestational
weight gain is predicted by obesogenic variants, so that the
apparent association between gestational weight gain and
child BMI may be explained by common genetic factors (28).
Several studies have also documented similar associations
between child BMI and the BMI of both parents (29–31),
which would be inconsistent with an in utero mechanism operating solely through the mother. Additional studies have
highlighted the clustering of obesity-related risk factors in
mothers and children and the role of the physical environment
in shaping both the mother’s and child’s BMI (32, 33). Any
of these shared genetic, behavioral, and environmental factors may help to explain the null within-family association of
gestational weight gain and prepregnancy BMI on child BMI
observed herein.
We found also that the relation between gestational weight
gain and child BMI appeared to be attenuated by birth weight
in both GEEs and fixed-effects analysis. This result is not
surprising given that gestational weight gain includes birth
weight and that birth weight was independently related to
child BMI. Birth weight does not appear to be a mediator
given that gestational weight gain and prepregnancy BMI
were not related to child BMI in fixed-effects analyses prior
to adjustment for birth weight. Instead, birth weight may
be capturing nonshared genetic or other environmental differences between siblings (28).
The major strength of this study was the longitudinal assessment of pregnancy characteristics and offspring size among
siblings, which made it possible to evaluate the impact of prepregnancy BMI and gestational weight gain on child BMI
while controlling for unmeasured stable maternal or family
traits. By conditioning on the family with fixed effects, our
aim was to generate results with minimal bias from common
family factors that may confound associations between prenatal exposures and child BMI, thereby eliminating a possible reason for conflicting results across prior studies.
Additionally, maternal height and weight prior to delivery
and child height and weight were measured in a standardized manner by health-care professionals instead of selfreported or reported from an outside source.
This study is also subject to several limitations. First, the
data collected during the early 1960s come from an older
cohort of women and children when excess gestational weight
gain and maternal obesity were less common. However, the
biologic mechanisms behind the relations between pregnancy
characteristics and offspring size should not change greatly
with time. In addition, as previously noted in the Discussion,
our results were similar to those based on more recent data
with higher rates of maternal obesity. There were also no data
Am J Epidemiol. 2011;174(10):1159–1165
1164 Branum et al.
available on possible changes in paternity across pregnancies
or paternal size. As in other studies, we relied on self-reported
prepregnancy weight that may be prone to underestimation.
Although the fixed-effects approach controls for unobserved
stable familial traits that vary between families, it does not
control for unmeasured, nonshared, or time-varying characteristics that occur within families (e.g., changes in environment
or maternal characteristics over time or genetic differences
between siblings). However, several important time-varying
maternal characteristics were observed and controlled (i.e.,
smoking, maternal age, socioeconomic status, parity), and only
an uncontrolled negative confounder—one that is related to
maternal BMI or gestational weight gain and child BMI in
opposing ways—could be responsible for the null association
observed herein. Finally, the CPP was not a representative
sample of pregnant women and, therefore, generalizability
is limited.
Our results contribute to an emerging picture of the complex
relations between maternal body size and gestational weight
gain and child’s size. The results of this study indicate that
the positive association between maternal weight gain prior
to and during pregnancy and child BMI may be confounded
by shared familial factors, such as genetic and behavioral
propensities. The fixed-effects approach contrasting sibling
outcomes for the same mother provides evidence that changes
in prepregnancy weight and gestational weight gain between
pregnancies are not associated with differences in offspring
BMI, at least at age 4 years. Although optimal gestational
weight gain may be important for other maternal and infant
outcomes, these results support an emphasis on genetic, behavioral, or postnatal environmental factors in determining
the child’s size, rather than an intrauterine mechanism. Future
studies should continue to evaluate the relative roles and
potential for intervention of important familial and environmental factors that may influence childhood BMI and obesity.
ACKNOWLEDGMENTS
Author affiliations: Infant and Women’s Health Statistics
Branch, Office of Analysis and Epidemiology, National
Center for Health Statistics, Centers for Disease Control and
Prevention, Hyattsville, Maryland (Amy M. Branum); Office
of Analysis and Epidemiology, National Center for Health
Statistics, Centers for Disease Control and Prevention,
Hyattsville, Maryland (Jennifer D. Parker); The Research
Institute at Nationwide Children’s Hospital, Center for
Biobehavioral Health, The Ohio State University College of
Medicine, Columbus, Ohio (Sarah A. Keim); and Office of
Epidemiology, Policy, and Evaluation, Maternal and Child
Health Bureau, Health Resources and Services Administration,
Department of Health and Human Services, Rockville,
Maryland (Ashley H. Schempf).
This paper was presented as an oral presentation at the
23rd Annual Meeting of the Society for Pediatric and Perinatal
Epidemiologic Research, Seattle, Washington, June 22–23,
2010, and at the 43rd Annual Meeting of the Society for
Epidemiologic Research, Seattle, Washington, June 23–26,
2010.
The findings and conclusions in this paper are those of
the authors and do not necessarily represent the views of the
National Center for Health Statistics, Centers for Disease
Control and Prevention.
Conflict of interest: none declared.
REFERENCES
1. Institute of Medicine and National Research Council.
Weight Gain During Pregnancy: Reexamining the Guidelines.
Washington, DC: The National Academies Press; 2009.
2. Oken E. Maternal and child obesity: the causal link. Obstet
Gynecol Clin North Am. 2009;36(2):361–377, ix–x.
3. Fisch RO, Bilek MK, Ulstrom R. Obesity and leanness at
birth and their relationship to body habitus in later childhood.
Pediatrics. 1975;56(4):521–528.
4. Whitaker RC. Predicting preschooler obesity at birth: the
role of maternal obesity in early pregnancy. Pediatrics. 2004;
114(1):e29–e36.
5. Oken E, Taveras EM, Kleinman KP, et al. Gestational weight
gain and child adiposity at age 3 years. Am J Obstet Gynecol.
2007;196(4):322.e1–322.e8.
6. Wrotniak BH, Shults J, Butts S, et al. Gestational weight
gain and risk of overweight in the offspring at age 7 y in
a multicenter, multiethnic cohort study. Am J Clin Nutr. 2008;
87(6):1818–1824.
7. Crozier SR, Inskip HM, Godfrey KM, et al. Weight gain in
pregnancy and childhood body composition: findings from the
Southampton Women’s Survey. Am J Clin Nutr. 2010;91(6):
1745–1751.
8. Olson CM, Strawderman MS, Dennison BA. Maternal weight
gain during pregnancy and child weight at age 3 years. Matern
Child Health J. 2009;13(6):839–846.
9. Simmons R. Perinatal programming of obesity. Semin Perinatol.
2008;32(5):371–374.
10. Hewitt JK. The genetics of obesity: what have genetic studies
told us about the environment. Behav Genet. 1997;27(4):
353–358.
11. Reilly JJ, Armstrong J, Dorosty AR, et al. Early life risk
factors for obesity in childhood: cohort study [electronic
article]. BMJ. 2005;330(7504):1357. (doi: 10.1136/bmj.
38470.670903.E0).
12. Skidmore PM, Cassidy A, Swaminathan R, et al. An obesogenic postnatal environment is more important than the
fetal environment for the development of adult adiposity:
a study of female twins. Am J Clin Nutr. 2009;90(2):
401–406.
13. Maes HH, Neale MC, Eaves LJ. Genetic and environmental
factors in relative body weight and human adiposity. Behav
Genet. 1997;27(4):325–351.
14. Faith MS, Pietrobelli A, Nuñez C, et al. Evidence for
independent genetic influences on fat mass and body mass
index in a pediatric twin sample. Pediatrics. 1999;104(1 pt 1):
61–67.
15. Rogers I. EURO-BLCS Study Group. The influence of
birthweight and intrauterine environment on adiposity and
fat distribution in later life. Int J Obes Relat Metab Disord.
2003;27(7):755–777.
16. Niswander KR, Gordon M. The Women and Their Pregnancies:
The Collaborative Study of the National Institute of Neurological
Diseases and Stroke. Philadelphia, PA: WB Saunders; 1972.
17. Hardy JB. The Collaborative Perinatal Project: lessons and
legacy. Ann Epidemiol. 2003;13(5):303–311.
Am J Epidemiol. 2011;174(10):1159–1165
Gestational Weight Gain and Child BMI in Siblings
18. National Center for Health Statistics. A SAS program for the
CDC growth charts. Atlanta, GA: Centers for Disease Control
and Prevention; 2010. (http://www.cdc.gov/nccdphp/dnpao/
growthcharts/resources/sas.htm). (Accessed November 2010).
19. Kleinman KP, Oken E, Radesky JS, et al. How should gestational weight gain be assessed? A comparison of existing
methods and a novel method, area under the weight gain curve.
Int J Epidemiol. 2007;36(6):1275–1282.
20. National Center for Health Statistics. 2000 Growth Charts for
the United States: Methods and Development. Ser 11, No. 246.
Hyattsville, MD: Centers for Disease Control and Prevention;
2002.
21. Myrianthopoulos NC, French KS. An application of the
U.S. Bureau of the Census socioeconomic index to a large,
diversified patient population. Soc Sci Med. 1968;2(3):
283–299.
22. Methodology and Scores of Socioeconomic Status. Washington,
DC: US Bureau of the Census; 1963.
23. Greenland S. Dose-response and trend analysis in epidemiology:
alternatives to categorical analysis. Epidemiology. 1995;6(4):
356–365.
24. Gilman SE, Gardener H, Buka SL. Maternal smoking during
pregnancy and children’s cognitive and physical development:
a causal risk factor? Am J Epidemiol. 2008;168(5):522–531.
25. Allison PD. Introduction to fixed effects methods. In: Fixed
Effects Regression Methods for Longitudinal Data Using SAS.
Cary, NC: SAS Institute, Inc; 2005:1–6.
Am J Epidemiol. 2011;174(10):1159–1165
1165
26. Susser E, Eide MG, Begg M. Invited commentary: the use of
sibship studies to detect familial confounding. Am J Epidemiol.
2010;172(5):537–539.
27. Schack-Nielsen L, Michaelsen KF, Gamborg M, et al.
Gestational weight gain in relation to offspring body mass
index and obesity from infancy through adulthood. Int J Obes
(Lond). 2010;34(1):67–74.
28. The NS, Adair LS, Gordon-Larsen P. A study of the birth
weight–obesity relation using a longitudinal cohort and sibling
and twin pairs. Am J Epidemiol. 2010;172(5):549–557.
29. Davey Smith G, Steer C, Leary S, et al. Is there an intrauterine
influence on obesity? Evidence from parent child associations
in the Avon Longitudinal Study of Parents and Children
(ALSPAC). Arch Dis Child. 2007;92(10):876–880.
30. Perez-Pastor EM, Metcalf BS, Hosking J, et al. Assortative
weight gain in mother-daughter and father-son pairs: an
emerging source of childhood obesity. Longitudinal study of
trios (EarlyBird 43). Int J Obes (Lond). 2009;33(7):727–735.
31. Lake JK, Power C, Cole TJ. Child to adult body mass index
in the 1958 British birth cohort: associations with parental
obesity. Arch Dis Child. 1997;77(5):376–381.
32. Cameron AJ, Crawford DA, Salmon J, et al. Clustering of
obesity-related risk behaviors in children and their mothers.
Ann Epidemiol. 2011;21(2):95–102.
33. Duarte CS, Chambers EC, Rundle A, et al. Physical characteristics of the environment and BMI of young urban children
and their mothers. Health Place. 2010;16(6):1182–1187.